Using a longitudinal correlational design, nine aims will be addressed. The first three aims examine the unique predictors of early lexical development and the potential mediating role that parental input may play in this predictive model. In Aim 1, we test whether there is incremental (i.e., significant prediction above and beyond other predictors) validity of (a) early child intentional communication, attention following, and object play, (b) early child motor ability and (c) time-varying child vocabulary comprehension in predicting growth rate of communicative word use. In Aim 2, we test whether there is incremental validity of (a) early child intentional communication, attention following, and object play, (b) attention to child-directed speech, and (b) parental variables in predicting growth rate of comprehension. In Aim 3, we test whether the association between early parental factors and the child's growth of communicative word use is mediated through the child's development of vocabulary comprehension. The next 3 aims involve testing longitudinal correlations that, if confirmed, are consistent with child effects on parental use of putative facilitative behaviors. In Aim 4, we test whether children's early frequency of intentional communication and attention following positively predict later frequency of parental linguistic mapping. In Aim 5, we test whether children's early object play positively predicts later parental synchronous nonobligatory talk. In Aim 6, we test whether children's early requesting positively predicts later parental tendency to use responsive strategies. The final 3 aims involve testing alternative models against key predictions of the model presented in Figure 1. In Aim 7, we test whether the association between early parent variables and later comprehension is greater than the association between early comprehension and later parent variables. In Aim 8, we test whether the association between early comprehension and later expression is greater than the association between early expression and later comprehension. In Aim 9, we test whether the association between early attention to child directed speech and later comprehension is greater than the association between early comprehension and later attention to child-directed speech. Mixed level modeling, multiple regression, and the modified Pearson-Filon statistic will be used to test the aims. Five years of funding are requested. PUBLIC HEALTH RELEVANCE: The incidence of autism is now estimated to be 1:150 births. Communicative word use by elementary school is a replicated predictor of social and employment outcomes in young adults with adults. Predicting variance in acquiring communicative word use may identify precursors that are appropriate early treatment targets.